Calculating within-month standard deviation from daily return data

I have imported a matrix with 2 columns into matlab. The 1st column is serial date number (not date string) and the 2nd column is daily returns from 1996-2014. I wish to calculate the standard deviation within each month, but the window lengths from month to month are different.

Please kindly help provide me the detailed scripts to do so as I'm new to matlab and I will need to do the same thing to 19 other institutions.

[Guillaume]

Unless you're willing to pay somebody, it's very unlikely you'll get a detailed script from here (Note I'm not offering).

Answer 1

[Andrei Bobrov]

[Y,M] = datevec(data(:,1));
[a,~,c] = unique([Y,M],'rows');
out = [a, accumarray(c,data(:,2),[],@(x)std(x))];

[Sophal]

Dear Andrei, you save my life. Although I don't really understand your script, it does work for the 1st institution. I'll continue for other 19 institutions. Many thanks and have a great day:)

[Sean de Wolski]

accumarray for the win!

[Anas Khalfi]

Dear Andrei, I think I have the same problem. I have financial data for different countries but because of business days, I don't have the same amount of values for each month. Could you give me a bit of explanations about your code because I tried to understand it to adapt it but I really don't have any idea about what you've done.

Answer 2

[Steven Lord]

Since this message was originally written we've introduced the timetable data type. I'd store the data in a timetable and call retime on the timetable, specifying 'monthly' as the value for the newTimeStep input and @std as the method input.